Venturing off the beaten path, Bryce Carey yesterday presented a poster at the 34th annual conference of the Society for Neuroscience about the interaction between an MCH class 1 protein and γ-secretase. The list of substrates for this key enzyme in amyloid production is growing continuously, but most of these substrates do not currently receive intense scrutiny for a possible role in the development of Alzheimer’s. This could change, however, if the notion becomes more firmly established that γ-secretase acts in AD pathogenesis not by a straightforward gain of function (i.e., Aβ production), but by a more subtle mix of partial gain and partial loss of function. Inklings of this trend pervade the field, and if it became more widely recognized, scientists will study the long list of γ-secretase targets with renewed interest to understand the broader role this proteolytic complex could have in AD and aging.

What’s with HLA, then? Carey, a technician in Dora Kovacs’s lab at Massachusetts General Hospital in Charlestown, began pursuing it when a sequence comparison showed that HLA-2A has an intramembrane domain shared by γ-secretase targets. But wait, you say—HLA is not a neuronal protein. True, it’s one of two classes of antigen-presenting membrane proteins, and its many varieties are traditionally thought to be expressed only on immune cells, where they present antigen to T cells to crank up an immune response. But a few years ago, Carla Shatz’s group at Harvard Medical School discovered to their surprise that developing, and indeed adult, mouse neurons express it, as well, and that lab is now studying class 1 MCH proteins for a possible function in the activity-dependent pruning that shapes the neonatal nervous system (see ARF related news story and ARF news).

In yesterday’s poster, Carey identifies the HLA-A2 as a substrate of α- and γ-secretase-mediated cleavage. He expressed HLA-A2 in CHO cells, and found that it undergoes ectodomain shedding when he also expressed ADAM-10, the leading candidate for the α-secretase role (see (see ARF related news story on Postina et al., 2004). This cleavage gives rise to a soluble piece and a membrane-anchored piece. The latter one then gets clipped further by γ-secretase, yielding a final soluble snippet that the cells quickly degrade, Carey’s poster suggests. Further experiments indicate that HLA-A2 forms a complex with β2-microglobulin, as indeed it normally does in lymphocytes. When Carey inhibited γ-secretase, less HLA was presented at the surface, hinting that γ-secretase cleavage might have to do with getting it there or be recycled properly.

If confirmed, one implication of this early work is that it could help explain why some γ-secretase inhibitor drugs interfere with T cell maturation (see ARF related news story). Furthermore, it might illuminate why some presenilin double knockout strains show changes in their thymocyte populations and mild autoimmune symptoms (see Tournoy et al., 2004). But most intriguing, perhaps, is speculation about what it might be doing in adult brain.—Gabrielle Strobel.


  1. The identity of substrates of γ-secretase is becoming more and more interesting—although the wealth of putative candidate substrates is also becoming somewhat disturbing. Scrutinizing all of them to define the exact physiological role (if any) of the processing of the "snipped-off" pieces of proteins will be a daunting task, but it is a "must" to understand their contributions to physiology and pathophysiology.

    It is surprising and even hard to believe that on the growing list of substrates (who keeps track?), only APP and the amyloid peptides are "offensive" substrates of mutant PS1 in causing early onset familial AD. Since even these most aggressive forms of FAD take decades to develop, one wonders why no other problems or diseases are associated with mutant PS1?

    As indicated rightfully by Gabrielle Strobel, the "gain- or loss-of-function" debate surrounding mutant PS1/γ-secretase will hopefully soon and finally be settled as a "gain- and loss-of-function" problem, i.e., gain as a proteinase by producing more Aβ42 and loss as a calcium ion regulating entity. This function is proposed to operate at synapses, although we do not understand how it is exerted and controlled.

    I remember as recently as in Philadelphia, and at previous meetings and in seminars, the skepsis when we tried to convey these notions. One cannot help but muse over the slow learning and fast forgetting in the AD-field.


    . Secretases as targets for the treatment of Alzheimer's disease: the prospects. Lancet Neurol. 2002 Nov;1(7):409-16. PubMed.

    . Capacitive calcium entry is directly attenuated by mutant presenilin-1, independent of the expression of the amyloid precursor protein. J Biol Chem. 2003 Jan 24;278(4):2484-9. PubMed.

    . Capacitative calcium entry induces hippocampal long term potentiation in the absence of presenilin-1. J Biol Chem. 2003 Nov 7;278(45):44393-9. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Immune Proteins Play Role in Brain Development and Remodeling
  2. Blowing a Cover: What Is T Cell Receptor, Key Immune Operative, Doing in Neurons?
  3. α-Secretase Returns to Center Stage
  4. New Orleans: Out Go Classic γ-Secretase Inhibitors, In Come More Dexterous NSAIDs?

Paper Citations

  1. . Partial loss of presenilins causes seborrheic keratosis and autoimmune disease in mice. Hum Mol Genet. 2004 Jul 1;13(13):1321-31. PubMed.

Further Reading


  1. . Immune signalling in neural development, synaptic plasticity and disease. Nat Rev Neurosci. 2004 Jul;5(7):521-31. PubMed.
  2. . HLA-A2 homozygosity but not heterozygosity is associated with Alzheimer disease. Neurology. 2002 Mar 26;58(6):973-5. PubMed.
  3. . No association of the HLA-A2 allele with Alzheimer's disease. Neurosci Lett. 2002 Dec 25;335(2):75-8. PubMed.